[jira] [Commented] (FLINK-8266) Add network memory to ResourceProfile
[ https://issues.apache.org/jira/browse/FLINK-8266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16339807#comment-16339807 ] ASF GitHub Bot commented on FLINK-8266: --- Github user asfgit closed the pull request at: https://github.com/apache/flink/pull/5170 > Add network memory to ResourceProfile > - > > Key: FLINK-8266 > URL: https://issues.apache.org/jira/browse/FLINK-8266 > Project: Flink > Issue Type: Improvement > Components: Cluster Management >Reporter: shuai.xu >Assignee: shuai.xu >Priority: Major > Labels: flip-6 > Fix For: 1.5.0 > > > In task manager side, it should allocated the network buffer pool according > to the input channel and output sub partition number, but when allocating a > worker, the resource profile doesn't contain the information about these > memory. > So I suggest add a network memory filed to ResourceProfile and job master > should calculate it when scheduling a task and then resource manager can > allocating a container with the resource profile. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (FLINK-8266) Add network memory to ResourceProfile
[ https://issues.apache.org/jira/browse/FLINK-8266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16325899#comment-16325899 ] ASF GitHub Bot commented on FLINK-8266: --- Github user shuai-xu commented on the issue: https://github.com/apache/flink/pull/5170 @tillrohrmann sorry, the conflict is resolved now > Add network memory to ResourceProfile > - > > Key: FLINK-8266 > URL: https://issues.apache.org/jira/browse/FLINK-8266 > Project: Flink > Issue Type: Improvement > Components: Cluster Management >Reporter: shuai.xu >Assignee: shuai.xu >Priority: Major > Labels: flip-6 > > In task manager side, it should allocated the network buffer pool according > to the input channel and output sub partition number, but when allocating a > worker, the resource profile doesn't contain the information about these > memory. > So I suggest add a network memory filed to ResourceProfile and job master > should calculate it when scheduling a task and then resource manager can > allocating a container with the resource profile. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (FLINK-8266) Add network memory to ResourceProfile
[ https://issues.apache.org/jira/browse/FLINK-8266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16324089#comment-16324089 ] ASF GitHub Bot commented on FLINK-8266: --- Github user tillrohrmann commented on the issue: https://github.com/apache/flink/pull/5170 Thanks for the contribution @shuai-xu. There are some merge conflicts. Could you please resolve them? > Add network memory to ResourceProfile > - > > Key: FLINK-8266 > URL: https://issues.apache.org/jira/browse/FLINK-8266 > Project: Flink > Issue Type: Improvement > Components: Cluster Management >Reporter: shuai.xu >Assignee: shuai.xu > Labels: flip-6 > > In task manager side, it should allocated the network buffer pool according > to the input channel and output sub partition number, but when allocating a > worker, the resource profile doesn't contain the information about these > memory. > So I suggest add a network memory filed to ResourceProfile and job master > should calculate it when scheduling a task and then resource manager can > allocating a container with the resource profile. -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (FLINK-8266) Add network memory to ResourceProfile
[ https://issues.apache.org/jira/browse/FLINK-8266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16292337#comment-16292337 ] ASF GitHub Bot commented on FLINK-8266: --- GitHub user shuai-xu opened a pull request: https://github.com/apache/flink/pull/5170 [FLINK-8266] [runtime] add network memory to ResourceProfile for the input and output memory of a task ## What is the purpose of the change This pull request adds a network memory field to ResourceProfile. So job master can set the network memory of a task according to the number of input channels and output sub partitions. ## Brief change log *(for example:)* - *The TaskInfo is stored in the blob store on job creation time as a persistent artifact* - *Deployments RPC transmits only the blob storage reference* - *TaskManagers retrieve the TaskInfo from the blob cache* ## Verifying this change This change can be verified by running ResourceProfileTest: ## Does this pull request potentially affect one of the following parts: - Dependencies (does it add or upgrade a dependency): (no) - The public API, i.e., is any changed class annotated with `@Public(Evolving)`: (no) - The serializers: (no) - The runtime per-record code paths (performance sensitive): (no) - Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Yarn/Mesos, ZooKeeper: (no) - The S3 file system connector: (no) ## Documentation - Does this pull request introduce a new feature? (no) - If yes, how is the feature documented? (not applicabled) You can merge this pull request into a Git repository by running: $ git pull https://github.com/shuai-xu/flink jira-8266 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/flink/pull/5170.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #5170 commit 93822bfac6a0794b2b2047e046dcef93c5313185 Author: shuai.xus Date: 2017-12-15T10:43:27Z [FLINK-8266] add network memroy to ResourceProfile for the input and output memory of a task > Add network memory to ResourceProfile > - > > Key: FLINK-8266 > URL: https://issues.apache.org/jira/browse/FLINK-8266 > Project: Flink > Issue Type: Improvement > Components: Cluster Management >Reporter: shuai.xu >Assignee: shuai.xu > Labels: flip-6 > > In task manager side, it should allocated the network buffer pool according > to the input channel and output sub partition number, but when allocating a > worker, the resource profile doesn't contain the information about these > memory. > So I suggest add a network memory filed to ResourceProfile and job master > should calculate it when scheduling a task and then resource manager can > allocating a container with the resource profile. -- This message was sent by Atlassian JIRA (v6.4.14#64029)
